Recently I am trying to new stored procedure MONREPORT to investigate
the performance issue. I get the following result and can't figure it
why.

--------------------------------------------------------------------------------
Monitoring report -
connection

--------------------------------------------------------------------------------
Database:
SAMPLE
Generated: 01/03/2012
16:52:11
Interval monitored:
30
-- Command options
--
APPLICATION_HANDLE:
All


================================================== ==============================
Part 1 - Summary of
connections


--------------------------------------------------------------------------------
APPLICATION TOTAL_ TOTAL_ ACT_COMPLETED TOTAL_WAIT
CLIENT_IDLE
# _HANDLE CPU_TIME ACT_TIME _TOTAL
_TIME _WAIT_TIME
--- ----------- ------------ -------- -------------
----------
-----------
1 29910 0 0 0
0 33038
2 29935 0 0 0
0 33039
3 29940 0 0 0
0 33038
4 29941 0 0 0
0 33038
5 29966 4594 7 34
4 24024
6 29969 0 0 0
0 33038
7 29980 0 0 0
0 33039
8 29985 534057 1040 3101
514 27604
9 29988 0 0 0
0 33039
10 29995 0 0 0
0 33039
11 30162 0 0 0
0 33039
12 30167 0 0 0
0 33039
13 30168 0 0 0
0 22037
14 30169 0 0 0
0 33039
15 30170 0 0 0
0 33039
16 30246 0 0 0
0
0
17 30252 0 0 0
0 33038
18 30253 0 0 0
0 33039
19 30254 0 0 0
0 33039
20 30255 756601 1462 4719
729 24925
21 30256 0 0 0
0 33039
22 30257 0 0 0
0 33039
23 30258 0 0 0
0 33038
24 30259 598194 1235 3808
690 25333
25 30296 739371 1355 4546
735 27102
26 30297 0 0 0
0 33038
27 30298 4241 3 23
2 23050
28 30299 642707 1431 4127
625 27347
29 30300 0 0 0
0 33039
30 30301 0 0 0
0 22037
31 30302 905992 1507 5727
880 28256
32 30303 0 0 0
0 33038


================================================== ==============================

The time interval is 30 seconds. According to the info center: This
monitor element records time spent waiting for the client to send its
next request. The value is given in milliseconds. So I am thinking
that value should be between 0 and 30,000ms.

From above output. some value is above the 30,000. I don't know why.

Any comments will be welcomed

Arthur

Hello,

On 04.01.12 14:54 , Arthurccc wrote:
Times reported could be slightly larger than the monitoring interval since
metrics are rolled up on a short time interval (approximately 10 seconds),
rather than instantaneously.

I opened a doc defect, which should result in an update of the documentation
with an example and an explanation.

--
Helmut K. C. Tessarek
DB2 Performance and Development
IBM Toronto Lab